We got money from the College this year to do some upgrades in the teaching labs.
one of the projects was a new computer for the xray powder diffraction system. I got all the software installed and then went to set up the machine, only to realize that the old machine has been around just long enough for me to forget that the interface card is a full-size PCI card. Every computer I've got that is newer than the current one doesn't fit the card either because Dell puts stuff in the way whenever they can, or there is a drive cage along the full height of the tower.
I can't replace the card, because I don't know if there's a smaller one, and this one cost $12k, so I don't really have budget for another one. I put the card back into the old computer (a 1.8GHz p4), upgraded the computer from 256MB of memory to 1GB, and then spent yesterday and today upgrading it to WinXP (from 98) and getting everything to work with the network login. Eventually, I may look into a big case and build a new computer.
During this whole process, I decided to use the (planned) new computer to replace the other computer in the lab, which controlled the older xray system. It was a much easier move (so far) than I thought it would be, as the hard drive on the computer I was replacing is dead (no head noise, no platter spinning noises, and a blinky LED on it's circuit board).
The new computer is set up with the analysis software and the search database, and everything there also works with the network, so that's good to go until somebody decides that they need the old machine to work, at which point if I haven't found a backup of the software (completely homemade, no backups in any obvious locations, the old backups probably having been eaten by a drive crash years ago), we'll decide whether we want to have the drive recovered, just scrap the whole works, or see if we can talk the ECE department into making it a senior design project, which is how we got the software in the first place.
During this whole process, I'm also printing posters. One of the grads from E's group emailed me a poster on Tuesday, with a note that there would be a few more for next week's conference. Yesterday I got another one, and today they sent me two more, with a note that there were another two coming. I replied by saying that I only had two or three posters worth of good paper left, so they might have to wait. It turned out that I had three posters worth, plus eleven inches of paper, which is just enough to finish the third poster without tripping the "out of paper" sensor. The guy who was doing the last two posters is willing to have them printed on bright white bond, so they'll get printed in the morning.
This is one of the reasons why work pays DoIT annually to make my email box hold 1GB (instead of the normal 200MB), as each poster arrives as an 8-20MB powerpoint file.
All the poster printing goes into my new hardware fund, and this batch is going to put another flat panel monitor into one of the teaching labs. It takes a conference or two to get anything good, as we basically only charge costs, so a 3'x4' poster costs either $30 or $45 (full-color posters - i.e. backgrounds too - cost more). They don't want me to charge more, and I won't charge less, as this way people think about the costs before they decide whether they're printing every random thing they have or not.
I, of course, get anything I want printed free, but that's the advantage of being the guy with the printer, and of being the guy who sends out the printer bills. Most of what I print out, however, is line art on plain paper (think floor plans and equipment drawings), so nobody really worries about it.
Tomorrow I'm teaching the TA that's going to be running the xray lab next week how the new setup works and I've got a guy from one of the Civil Engineering research groups coming over to use one of the microscopes to take pictures of nanoparticles.
Now, off to think about what to make for dinner, not knowing when teeka
will be done teaching class.